ó áp7]c@sÒddlZddlmZddlmZddlmZmZddl m Z ddl m Z dddedd„Zddddd „Zddddd „Zddddd „Zddddd „ZdS( iÿÿÿÿN(t arima_process(tprefix_dtype_map(t _get_epsilontapprox_fprime_cs(tfind_best_blas_typei(t_arma_innovationscCs•tj|ƒ}|jdk}|r=|dd…df}ntj|dkrUgn|ƒ}tj|dkrvgn|ƒ}|j\}}tjd| f} tjd|f} |dkrît|||tj|ƒgƒ\}} } nt|} tj |d| ƒ}tj |d| ƒ}tj |d| ƒ}| |ƒj ƒ}t t |dƒ} t t |dƒ}t t |dƒ}t j| | d|d|ƒ|}| | | |ƒ\}}||||||ƒ\}}tj|ƒ}|rü|d }ng}x_t|ƒD]Q}tj||dd…|f|||ƒƒ}|j|rY||n|ƒqWtj|ƒj}|r‹|jƒ}n||fS( s| Compute innovations using a given ARMA process Parameters ---------- endog : ndarray The observed time-series process, may be univariate or multivariate. ar_params : ndarray, optional Autoregressive parameters. ma_params : ndarray, optional Moving average parameters. sigma2 : ndarray, optional The ARMA innovation variance. Default is 1. normalize : boolean, optional Whether or not to normalize the returned innovations. Default is False. prefix : str, optional The BLAS prefix associated with the datatype. Default is to find the best datatype based on given input. This argument is typically only used internally. Returns ------- innovations : ndarray Innovations (one-step-ahead prediction errors) for the given `endog` series with predictions based on the given ARMA process. If `normalize=True`, then the returned innovations have been "whitened" by dividing through by the square root of the mean square error. innovations_mse : ndarray Mean square error for the innovations. iNtdtypetarma_transformed_acovf_fasttarma_innovations_algo_fasttarma_innovations_filtertsigma2tnobsgà?(tnptarraytndimtNonet atleast_1dtshapetr_RRtasfortranarraytitemtgetattrRRt arma_acovftrangetappendtvstacktTtsqueeze(tendogt ar_paramst ma_paramsR t normalizetprefixtsqueezedR tk_endogtartmaRt_RRR Rtacovftacovf2tthetatvtv05tutitu_i((sKlib/python2.7/site-packages/statsmodels/tsa/innovations/arma_innovations.pytarma_innovations sP"!! *   !c Cs1t|d|d|d|d|ƒ}tj|ƒS(s¢ Compute loglikelihood of the given data assuming an ARMA process Parameters ---------- endog : ndarray The observed time-series process. ar_params : ndarray, optional Autoregressive parameters. ma_params : ndarray, optional Moving average parameters. sigma2 : ndarray, optional The ARMA innovation variance. Default is 1. prefix : str, optional The BLAS prefix associated with the datatype. Default is to find the best datatype based on given input. This argument is typically only used internally. Returns ------- loglike : numeric The joint loglikelihood. RRR R (tarma_loglikeobsR tsum(RRRR R tllf_obs((sKlib/python2.7/site-packages/statsmodels/tsa/innovations/arma_innovations.pyt arma_loglikeescCs tj|ƒ}tj|dkr'gn|ƒ}tj|dkrHgn|ƒ}|dkrŠt|||tj|ƒgƒ\}}}nt|}tj|d|ƒ}tj|d|ƒ}tj|d|ƒ}||ƒjƒ}t t |dƒ}|||||ƒS(sÌ Compute loglikelihood for each observation assuming an ARMA process Parameters ---------- endog : ndarray The observed time-series process. ar_params : ndarray, optional Autoregressive parameters. ma_params : ndarray, optional Moving average parameters. sigma2 : ndarray, optional The ARMA innovation variance. Default is 1. prefix : str, optional The BLAS prefix associated with the datatype. Default is to find the best datatype based on given input. This argument is typically only used internally. Returns ------- loglikeobs : array of numeric Array of loglikelihood values for each observation. Rtarma_loglikeobs_fastN( R R RRRRtascontiguousarrayRRRR(RRRR R RR%tfunc((sKlib/python2.7/site-packages/statsmodels/tsa/innovations/arma_innovations.pyR/ƒs!! * csž|dkrgn|}|dkr*gn|}t|ƒ‰t|ƒ‰‡‡‡fd†}tj|||f}t|ddt|ƒƒ}t|||ƒS(s± Compute the score (gradient of the loglikelihood function) Parameters ---------- endog : ndarray The observed time-series process. ar_params : ndarray, optional Autoregressive coefficients, not including the zero lag. ma_params : ndarray, optional Moving average coefficients, not including the zero lag, where the sign convention assumes the coefficients are part of the lag polynomial on the right-hand-side of the ARMA definition (i.e. they have the same sign from the usual econometrics convention in which the coefficients are on the right-hand-side of the ARMA definition). sigma2 : ndarray, optional The ARMA innovation variance. Default is 1. prefix : str, optional The BLAS prefix associated with the datatype. Default is to find the best datatype based on given input. This argument is typically only used internally. Returns --------- score : array Score, evaluated at the given parameters. Notes ----- This is a numerical approximation, calculated using first-order complex step differentiation on the `arma_loglike` method. cs*tˆ|ˆ |ˆˆˆ!|ˆˆƒS(N(R2(tparams(Rtptq(sKlib/python2.7/site-packages/statsmodels/tsa/innovations/arma_innovations.pyR5×sg@N(RtlenR RRR(RRRR R R5tparams0tepsilon((RR7R8sKlib/python2.7/site-packages/statsmodels/tsa/innovations/arma_innovations.pyt arma_score¯s"  csž|dkrgn|}|dkr*gn|}t|ƒ‰t|ƒ‰‡‡‡fd†}tj|||f}t|ddt|ƒƒ}t|||ƒS(sÔ Compute the score per observation (gradient of the loglikelihood function) Parameters ---------- endog : ndarray The observed time-series process. ar_params : ndarray, optional Autoregressive coefficients, not including the zero lag. ma_params : ndarray, optional Moving average coefficients, not including the zero lag, where the sign convention assumes the coefficients are part of the lag polynomial on the right-hand-side of the ARMA definition (i.e. they have the same sign from the usual econometrics convention in which the coefficients are on the right-hand-side of the ARMA definition). sigma2 : ndarray, optional The ARMA innovation variance. Default is 1. prefix : str, optional The BLAS prefix associated with the datatype. Default is to find the best datatype based on given input. This argument is typically only used internally. Returns --------- scoreobs : array Score per observation, evaluated at the given parameters. Notes ----- This is a numerical approximation, calculated using first-order complex step differentiation on the `arma_loglike` method. cs*tˆ|ˆ |ˆˆˆ!|ˆˆƒS(N(R/(R6(RR7R8(sKlib/python2.7/site-packages/statsmodels/tsa/innovations/arma_innovations.pyR5sg@N(RR9R RRR(RRRR R R5R:R;((RR7R8sKlib/python2.7/site-packages/statsmodels/tsa/innovations/arma_innovations.pyt arma_scoreobsßs"  (tnumpyR tstatsmodels.tsaRt statsmodels.tsa.statespace.toolsRtstatsmodels.tools.numdiffRRtscipy.linalg.blasRtRRtFalseR.R2R/R<R=(((sKlib/python2.7/site-packages/statsmodels/tsa/innovations/arma_innovations.pyts  Z  +  /